How to Plan an MVP
by Michael Seibel
This article aims to collect the information in Y combinator for our entrepreneurs and give it as raw information. The fourth of these articles, which we named as startup 101, is “how to plan an MVP”.
Let’s begin 😊
The minimum eligible product is only those with sufficient features, the product offered to the first users, the product to be requested for feedback (wiki). In the lesson, the ridiculous thing is that you can only give the first step and give them a value, first of all, imagine 😉.
The first users you’re going to MVP with are essential to who you are, because these are the people, you’re solving the sizing problem of. You need to talk to these people before you write any more code.
Before releasing the product, the main goals of the startups are very simple. These:
Quickly Publish (MVP): You need to publish the first product quickly, even if it’s bad.
Find First Users: You don’t need to appeal to everyone, just find first-stage users.
Get Feedback: Do not think that the feedback you will receive from the user in the first version is meaningless before the full version is released. The product you have in mind may be a great thing for you, but perhaps it may be a product that these users will never/will ever use/need.
Iteration: Update your product constantly, iterate constantly, constantly improve.
There are basically two types of MVPs based on your initiative and solution: Lean and Heavy MVP.
It is the category with many initiatives. You have to MVP very quickly, it should be weeks, not months. Many startups do this with a landing page and a spreadsheet. Publish with limited feature so you just need to solve the most basic problem of users. Only serve a small number of users and iterate over them. Your product’s vision should encompass everyone, but a small number of users is enough for an MVP. MVP should not be seen as something very special. For example, when Airbnb first came out, it did not even have a payment system and a mapping feature, it was coded as a shambles. The original version of Twitch was Justin.TV, there was only a chat box where you could watch Justin’s life in very bad resolution and talk to other viewers. These later became billion-dollar companies, but at first they were very bad.
Complex (Heavy) MVP
There is no such term, but a name given to describe the event. In some cases MVP may not be very fast. These can be solutions for areas with intense rules and laws such as insurance and banking, or solutions for areas that require intensive labor such as drones, rocket technologies, biotechnology. Initiatives like this need to start with a website that tells you what you’re doing, so it’s easier to talk to people and you’ll give them something to refer to.
With the launch of the product, you will meet your customers for the first time. This should not be exaggerated either. For example, when Facebook, Google or Twitter went live, nobody remembers. With MVP, you need to get your first customers and continue with them by constantly improving your solution.
Tips for making a quick MVP:
- Set a Time: Set a time, such as three weeks, and do what you can within that time. Thus, you will not have time to put all the features on the product you want it to be, and a very simple product will emerge.
- Determine the Features: Unless you specify the features that will be in your product in writing at the beginning, you will have the opportunity to constantly change this 3-week product to a 3-month product; You can also convert a 3-month product into a 3-year product, and in this way, you will be out of the definition of MVP.
- Limit Features: The features you can give in a limited time are also limited. Limit features instead of constantly delaying time. Even waive some important features if necessary. When you introduce a product, the development process will continue with the momentum you gain. In order to capture exactly this momentum, put out the minimum suitable product as soon as possible.
- Don’t fall in love with your MVP: Never forget that it’s just a baby step.